Audio engineer explains NPR’s signature sound

C

combatentropy

Veteran
A top audio engineer explains NPR’s signature sound:
- the Neumann U87 microphone ($3,200 street)
- the bass roll-off switch on
- no further processing
- speaking just 6 to 12 inches from the microphone
- but not speaking straight into it, instead "at a diagonal, sort of off to the side of the microphone"
- a studio with little reverb or background noise

Nothing unusual, just the diligent application of simple, traditional techniques. Well, a $3,000 microphone may be unusual for many of us on DVXUser, but those large-diaphragm condenser microphones are the norm in radio stations, and that particular model may be the "most widely used studio microphone in the world," according to the product description on Amazon.
 
I think the key here is the high pass / bass roll off filter as if you get a decent cardioid condenser mic it will also sound good.

Any cardioid mic used up close will have a huge bass lift or proximity effect so using a high pass filter will help this. A dynamic mic with a high pass filter at the desk end can also sound good and the shure SM7 and EV RE20 are great mics but radio stations tend to not put the filter on so you get that very bass heavy effects once they compress it to oblivion.

Personally I sometimes also use an omni directional mic for close voice work as the gain structure has to be a lot lower so background noise tends not to be a problem.

I have been lucky to use the U87 and other high end mic's such as the AKG 414 for vocals and voice over work but at the moment find that the Rode NT2000 is a very good budget mic as is the SE X1. I have both and use them all the time so you don't have to spend $3,000.

As most of these mic's also have a -10db pad setting that can also help to reduce overload if used very close for vocals as it is usually in circuit before any in mic amplification.
 
Last edited:
Richard Crowley said:
Yeah, people who operate on public money can spend $3000 on an announce microphone in a radio studio because they don't have to answer to anybody. Meanwhile radio stations who operate in the Real World use more realistic microphones like ElectroVoice RE-20 large diameter dynamic microphones, etc. which are much more likely "the norm in radio stations". And at an order of magnitude more sensible price. NPR, PBS, et.al. are perfect examples of the old adage: "You can solve any problem by throwing enough money at it." Bah, humbug. Not terribly "clever" in my book.
Click to expand...

Bah humbug is correct. Is there something wrong with using great (expensive) gear? All kinds of productions use high end mics and if I had a permanent studio which would operate daily for years, a $3,000 mic would seem like a worthwhile investment. Given it produced a better product than the usual suspects. Since some folks seem think there is a "signature sound", I guess it is working?

Maybe NPR purchased these mics back when they had more public funding but nowdays they do not get hardly any public (government) money.
 
Bassman2003 said:
Since some folks seem think there is a "signature sound", I guess it is working?
Click to expand...

NPR has a very clean, full sound. It's one of the best production qualities I've heard on the radio. As Gary pointed out, not properly utilizing the mic's built-in shape tools and then processing the signal into oblivion leads to a bassy signal. That may be signature of commercial FM, but it's also one reason I particularly do not enjoy listening to commecial FM.

If you're ever in DC for a few days, call NPR and ask when they do their studio tours. It's a beautiful facility. It almost has to be, for a national radio network that does the kinds of production work they do. As for their signature sound, clean and simple are two things that matter. Start with a great mic, play to it correctly, and skimp on the post processing. There are many things in this world that can benefit from that practice.

The other thing to note is that NPR sending out very clean source material can help account for differences in FM transmissions on the local level. I do a lot of Interstate driving, often Knoxville to Chattanooga and Knoxville to Memphis. I can drive to Chattanooga during All Things Considered and hear a stunning difference when I switch from Knoxville's station to Chattanooga's. WUOT's broadcast of ATC sounds very pure, close to the original, but once I switch to WUTC I lose low end and get lots of upper mids. It doesn't sound nearly as good as what I get from Knoxville.

So if you want to maintain as best as possible the integrity of your national programming, and you know that different local stations have different processing chains, aren't you better off sending it out as clean and unprocessed as you can?
 
U87 is a great mic but is not the end of the story these days, back in the 80's I saw engineers automatically dial-ing in +15db at 10khz before it hit the vocal compressor and as I said I now personally have far lower cost mic's that sound just as good for most applications.

Also bear in mind that a $3,000 mic into a $25,000 + mixing board may sound great but a $3,000 mic into a $3,000 mixer may have you just wasting $3,000 especially if it has Yamaha written on it.

Even our $250,000 + Calrec mixers at work don't sound like the old analogue boards tat the U87 was designed to work with.
 
no further processing
- Maybe in the studio, but downstream there's surely an Optimod or other type of processor(s).
 
I think that's the point made above, send it out clean so that the station doesn't double process the audio. In most modern stations around here, there is a processor for the mics, and nothing until it hits the STL or transmitter (depending on location) where it definitely should be hitting some kind of processing to make sure it doesn't go too "loud".

On our radio stream I use an Orban Optimod PC (card and software) before it goes to the stream encoder. Careful crafting of the source signal is required for the compressed stream. If not you can "eat bits" on frequencies that can't really be sent. Same goes for the digital radio signal. Analog processing was often used to make the analog transmission sound "louder", louder was better and also made the transmitter reach a little farther (cheated the power measurements).
 
I was a bit surprised at the engineers comment that the roll off was at around 250Hz, which is way above most. He was wrong, but not in the direction I was expecting. In the cardioid position the -3dB point is around 500Hz! The reasoning from the company is that the roll off isn't a handling noise fix as it is in most small condensers, it's to compensate for the proximity effect of close micing.

A few things though that should be mentioned. This is only for the people IN the studio. When you hear interviews on Fresh Air or All Things Considered the person being interviewed is almost never at the studio and rarely being interviewed by the host. A line producer calls and asks the questions, a local stringer records the interview end while the interviewee is talking on the phone and then the stringer sends the files in, they edit for content and then write questions that better fit the answers. The host records the new questions and then that gets edited in with the edited answers and gets played as a "live" interview. The stringers are recording in people offices etc. and are using whatever mic they normally use. I was mostly using an Oktava Hyper.

The other point is that out side of local programing everything on NPR has been compressed to an MP3 (though they may have now moved to MP4) file and uploaded to a central depository (PRX). So the added benefit of not being bassy is the encoding comes out better.

Having been involved in one way or another with a bunch of NPR, PBS stations over the years I have yet to see one swimming in cash or full of well paid staff. The union positions are probably the best paid positions there outside of upper management, whom I assume like all upper management I have ever met, are probably paid more than I would think they are worth. In the general scheme of things though your local NPR affiliate probably produces more local content that most of the other stations, not counting sports. Most music stations are programed from central locations in other states, a lot of "talk" radio is actually being run from remote locations in other states. There are only a few, usually small niche, stations that actually produce real local content.

I don't really think this is a political issue. PBS started in 1954 as NET and was funded by the Ford Foundation. It was entirely privately funded till 1967 when Ford started cutting back and congress set up CPB to help keep noncommercial broadcast alive. It's kind of ironic since originally all broadcast was supposed to be non commercial. Other than some economic hardship the biggest impact if CPB disappeared would probably be the loss of local programing since that is one of the things they enforce.

In terms of expensive mics I think you have to look back to when most were probably bought (early 1960's). There wasn't a lot of options. I also suspect that for field use they had a bunch of the radio versions of the Nagra (I even have one). Something that would come off as way over priced today. Today though most of the field work is done on some variation of a Zoom handheld. I know a independent producer and she uses a Schoeps. Seems crazy expensive, right? She bought it in the early 70's because that was the mic that everyone used for field work because it could be powered from the mixer with T power. When $$$ options were the only decent ones people found ways to get the $$$ option. It helped that they were not dropping close to a grand a month on things like internet, cel phones, cable, computers etc. since those didn't exist yet.
 
Rick R said:
no further processing
- Maybe in the studio, but downstream there's surely an Optimod or other type of processor(s).
Click to expand...

At the local member station, sure, but not leaving NPR. From the article linked above:

Fox: Well, speaking to the public radio community, one of the things that we don’t do is process our signal heading to the stations. We keep it as pure as possible. This has been up for some debate for the longest time, but as we currently stand we don’t compress; we don’t alter our signal leaving the studio to the satellite and hitting the stations.

We know that our over 600 stations and all their listeners have different needs. So we try to give the purest sound to the stations so that they can manipulate it for their market, with varying degrees of success. We have talked about, especially with our newscast unit, maybe starting to slide a little bit of compression into that. We haven’t done it yet, and I don’t know if we will, but giving that pure signal via the satellite to our stations is probably the last element to the sauce.
Click to expand...
 
I doubt they put their content on their servers as mp3, mp2 is about the most compressed you dare go before you start to hear it down the chain and into digital transmission. Our local PBS had everything as wave files nearly 10 years ago, we have everything as wave files, etc. I did mess around with 192kbps MP3 and it sounded marginal after it hit the stream encoder. The recommended storage was to go full 44.1khz 16 bit wave or go mp2 compression if you really needed to save space. At the cost of hard drives and servers, it just doesn't make sense to use anything less than CD quality wave files for broadcast. I can tell you whenever we are playing music that originated from iTunes (any compression), you can hear the artifacts after it gets blown back up to PCM, and compressed back down to aac+ on the stream, most of them can be identified in the studio just listening to the mix.
 
You are right about the MP3 it is MP2. From PRX's submission page "reminder, PRX only accepts MP2s".

I didn't see it listed but they also now require you to hit certain levels on a LUFS meter.

Your local station for local broadcast may keep everything as a wav file but everything not local is downloaded from PRX and that is all MP2.
 
And here's the real rub... While it's nice having a $250,000 audio console and $3600 microphones. As long as the story is compelling people would listen. Even if they used a Marantz cassette deck and a 635a mic.
 
David W. Jones said:
And here's the real rub... While it's nice having a $250,000 audio console and $3600 microphones. As long as the story is compelling people would listen. Even if they used a Marantz cassette deck and a 635a mic.
Click to expand...

Field reporters about 10 years ago were using MiniDisc recorders.
 
And before that indeed the Marantz portable cassette recorder was used a lot. It was great except the clutch failed a lot.
 
It was a Uher in my days with their plastic reporter mic. :lipsrseal

Totally agree and remember Betacam was cassette quality with no noise reduction and then Beta SP had dolby C.

I am now back working on mainstream TV news after a 25 year break with £250,000 Calrec boards but hear more and more distorted and bad audio that is just churned out regardless of quality.

All of the press conferences at the recent European football championships are done with coms talkback mic's that pop and splat all over the place.
 
Last edited:
You must log in or register to reply here.
Back
Top